Recovering Articulated Object Models from 3D Range Data

TL;DR

This paper presents an unsupervised algorithm that recovers articulated object models from 3D range data by segmenting meshes into rigid parts and identifying the skeleton, demonstrated on real-world datasets.

Contribution

The method automatically decomposes complex articulated objects into rigid parts and infers their skeletons from multiple 3D configurations without supervision.

Findings

Successfully recovered a 15-part human puppet model from 7 configurations.

Effectively identified a 4-part model of a flexible arm with non-rigid deformation.

Efficiently performed optimal segmentation despite densely connected graphical models.

Abstract

We address the problem of unsupervised learning of complex articulated object models from 3D range data. We describe an algorithm whose input is a set of meshes corresponding to different configurations of an articulated object. The algorithm automatically recovers a decomposition of the object into approximately rigid parts, the location of the parts in the different object instances, and the articulated object skeleton linking the parts. Our algorithm first registers allthe meshes using an unsupervised non-rigid technique described in a companion paper. It then segments the meshes using a graphical model that captures the spatial contiguity of parts. The segmentation is done using the EM algorithm, iterating between finding a decomposition of the object into rigid parts, and finding the location of the parts in the object instances. Although the graphical model is densely connected, the object decomposition step can be performed optimally and efficiently, allowing us to identify a large number of object parts while avoiding local maxima. We demonstrate the algorithm on real world datasets, recovering a 15-part articulated model of a human puppet from just 7 different puppet configurations, as well as a 4 part model of a fiexing arm where significant non-rigid deformation was present.